Key points:
Rather than replacing student thinking, when teachers design and guide AI experiences, the technology is most often used to deepen critical thinking and strengthen instruction, according to new insights from SchoolAI.
The research, AI isn’t replacing thinking: Teachers are using SchoolAI to deepen it and boost engagement, offers educators, school leaders, and policymakers large-scale evidence of how AI is actually being used in classrooms.
The report analyzed more than 23,000 teacher-created SchoolAI ‘Spaces’ used during the 2024-25 school year. These Spaces span English language arts, math, science, and social studies across elementary, middle, and high school classrooms. To answer the question of AI’s impact on student learning, we must first understand how it’s being used in the classroom. This study examined what teachers built and how students were asked to think when AI was involved.
Across subjects and grade levels, the data shows that higher-order thinking appears far more often than simple recall. Seventy-three percent of lessons require conceptual understanding, while 59 percent ask students to analyze information, and 58 percent ask them to evaluate ideas or make judgments. More than 75 percent of AI-supported lessons remain grounded in core academic curriculum, showing that teachers are extending familiar instruction rather than replacing it.
“There has been a lot of speculation about what AI might do to learning,” said Caleb Hicks, founder and CEO of SchoolAI. “This research gives educators, leaders, and policymakers something far more useful: evidence of what teachers are actually doing. When teachers design the experience and set clear expectations, AI becomes a way to push students toward deeper reasoning, analysis, and judgment. It supports rigorous thinking rather than replacing it, which is why AI can be a valuable tool for classroom learning.”
The study also highlights how teachers are using AI to create interactive, engaging learning experiences at scale while maintaining academic rigor. In science classrooms, roughly 25 percent of Spaces encourage open-ended investigation, while role-play and simulation appear in 18-20 percent of reading and social studies lessons.
At the same time, teachers recognize the importance of boundaries in responsible AI use. Teachers reinforce learning instead of simply looking up answers by designing experiences that push students toward deeper reasoning, not shortcuts.
“This study was designed to look at practice, not predictions,” said Cynthia Chiong, principal research scientist at SchoolAI. “We wanted to understand the kinds of thinking teachers are intentionally asking for when AI is involved. The findings offer concrete evidence of how teacher-led design shapes meaningful and responsible use of AI in real classrooms.”
Together, the findings challenge common fears about AI undermining learning. The research shows that when teachers lead the design, AI can strengthen critical thinking, increase engagement, and support responsible instruction across classrooms.
This press release originally appeared online.
